This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 4399 - Use XML Schema Definition Language (XSD), not XML Schema, as name of language
Summary: Use XML Schema Definition Language (XSD), not XML Schema, as name of language
Status: CLOSED FIXED
Alias: None
Product: XML Schema
Classification: Unclassified
Component: Structures: XSD Part 1 (show other bugs)
Version: 1.1 only
Hardware: Macintosh All
: P2 normal
Target Milestone: ---
Assignee: C. M. Sperberg-McQueen
QA Contact: XML Schema comments list
URL:
Whiteboard:
Keywords: resolved
Depends on:
Blocks:
 
Reported: 2007-03-19 23:40 UTC by C. M. Sperberg-McQueen
Modified: 2008-03-08 15:43 UTC (History)
0 users

See Also:


Attachments

Description C. M. Sperberg-McQueen 2007-03-19 23:40:06 UTC
Every published draft of the Structures specification has described Structures
as specifying (part of) the XML Schema definition language.  And every
draft has used XML Schema (with or without a 1.0 or 1.1) as the short
name of the language. 

The short names "XML Schema" and "XML Schema language" have predictably
led to confusion.  In writing, one can say that "the XML Schema language
is one of many XML schema languages", and even a moderately careful 
reader may follow the train of thought.  But try to say it aloud, and
you will be forcibly struck by the futility of relying on the difference 
between uppercase S and lowercase S to carry the difference between 
a proper noun (the name of our langauge) and a common noun phrase 
(which denotes an infinite class of languages including ours).  We 
would do ourselves, and everyone who writes, speaks, reads, or listens 
to people speaking about schema languages a favor if we made it easier 
to distinguish references to our language from references to schema 
languages in or for XML in general.

The name "XML Schema" has also given the Schema WG, and the W3C
in general, a reputation for arrogance; since the name seems to
convey the idea that we believe there is only one XML schema language worth
talking about, namely "the" XML Schema language.  A number of people
otherwise well disposed towards W3C (although not particularly well
disposed to the XML Schema WG, being no particular fans of our work) 
have publicly and privately expressed irritation at the name "XML Schema" 
and many refuse to use it; the alternative soubriquets they have bestowed
on the language vary, but "the W3C XML Schema [or schema] language",
abbreviated WXS, is not uncommon.

I think we would do well to listen to the complaint, since it seems
well founded. 

So I propose that we change our usage to give our language a
name which is not quite so readily confused with generic references
to XML schema languages, and adopt the new name in 1.1.

One possible new name:  call the language not the "XML Schema
description language" but the "XML Schema Description Language",
XSD or XSDL for short; change the main title line of the
spec from "XML Schema 1.1: Structures" to "XML Schema Definition
Language (XSD) 1.1:  Structures"; change references to "XML
Schema" in the text to "XSD" or to some other phrase, as
appropriate.

Other names and acronyms or initialisms are also possible; it
is less important to me that we use a particular name or
short form than that we avoid using what sounds like a common
noun phrase as a name or short form.  (The acid reception in
some quarters to the CSS spec called simply "Selectors" is
a good example of the reaction to our spec which I would like
to begin to change.  It could be elevated to a general rule:
if you do not wish to come across as hopelessly arrogant, 
avoid capitalizing a common noun phrase describing your
spec and giving your language that name.)  

One alternative to "XML Schema Definition Language" mentioned
when we discussed this topic in New Orleans was "XML Schema
Language"; unfortunately, I think I heard the other day that the 
initialism "XSL" is spoken for.

Note that the issue I'm raising applies equally to Structures
and Datatypes, but I don't expect separate discussion to be
needed so I am not opening a separate issue for tracking this 
question in the context of Datatypes.
Comment 1 Noah Mendelsohn 2007-03-21 14:12:27 UTC
I'm sympathetic to the spirit of what's proposed.  Michael suggests:

> One possible new name: call the language not the
> "XML Schema description language" but the "XML
> Schema Description Language", XSD or XSDL for
> short; change the main title line of the spec from
> "XML Schema 1.1: Structures" to "XML Schema
> Definition Language (XSD) 1.1: Structures"; change
> references to "XML Schema" in the text to "XSD" or
> to some other phrase, as appropriate.

I'm a bit surprised, as this seems only marginally less generic, and by the way just a bit clunkier.  My intuition is that with a name like this, people will still informally call it the XML Schema Language anyway.  I do understand that it goes some way toward licensing the use of the "D" in "XSD" as more than an artifact of the conventional file extension of our schema documents.

FWIW: I'm a little more in favor of:

Title: W3C XML Schema Definition Language

First of all, definition feels a bit closer to the mark to me than description, but I wouldn't burn a lot of time on that if there's disagreement.

I think this also supports the usage I've adopted in polite company, which is to call it the "W3C XML Schema Language" or for short the "W3C Schema Language" when there's likely to be either ambiguity or sensitivity.   The word "definition" also supports XSD as an informal shorthand for those who prefer it, though I find that I don't.  Still, I wouldn't particularly object to:

Title: W3C XML Schema Definition Language(XSD)

which would go some way toward making formal what people are doing anyway.  

Noah
Comment 2 Dave Peterson 2007-03-21 14:53:46 UTC
(In reply to comment #1)

> FWIW: I'm a little more in favor of:
> 
> Title: W3C XML Schema Definition Language

> Still, I wouldn't particularly object to:
> 
> Title: W3C XML Schema Definition Language(XSD)

All well and good, for the XML language used to define these schemas.  But we expect to see "born binary" schemas that are not attached to a particular defining document.  Are we not drifting toward ignoring the schemas themselves?  What do we call *them*?  "W3C XML Schemas ("XS)"?
Comment 3 Noah Mendelsohn 2007-03-21 22:29:10 UTC
I think the proposals are OK with respect to born binary.  My suggested title refers to a schema definition language.  I think that can be broadly interpreted as covering both the transfer syntax and also the more abstract specification of a schema (which you refer to as born binary, though I note that our specification really just provides abstractions, not any particular binary, hex, octal, etc. representation.)  Anyway, I'm usually sensitive to overlooking the components/"binary" side of things, and my antennae aren't twitching on this particular one.

Noah
Comment 4 C. M. Sperberg-McQueen 2007-05-25 22:32:03 UTC
A proposal to change the name of the spec and the name of the language
it defines was adopted by the WG on 11 May.  A summary may be found on
the IG list archive (member-only link) at
http://lists.w3.org/Archives/Member/w3c-xml-schema-ig/2007May/0002.html

Accordingly, I'm changing the keyword for this issue from needsDrafting
to decided.
Comment 5 C. M. Sperberg-McQueen 2007-06-18 14:15:58 UTC
The proposal adopted by the Working Group has now been integrated into
the status quo documents on the server; accordingly, I'm changing the
status to 'RESOLVED'.